Emergency Department Utilization by HIV-Infected Patients
The 2009 and 2010 NHAMCS surveys together estimated that there were 217 261 467 ED visits (95% CI 195 109 524, 239 413 410) by individuals aged ≥ 13 years in the USA. Of these patient ED visits, 1 192 535 (95% CI 934 410, 1 450 660) were documented to be visits by HIV-infected patients, corresponding to a visit proportion of 5 in 1000 ED visits (95% CI 4, 7 per 1000 visits). Numbers of ED visits and their corresponding visit proportions of all ED visits for known HIV-infected individuals by demographic characteristics are summarized in Table 1. The highest proportions of ED visits for HIV-infected individuals were found for those who were aged 45–54 years, were male, were of Black race, had public medical insurance as payment type, and were living in metropolitan areas and the northeast region. Using the US Census estimate of the civilian noninstitutionalized population aged ≥ 13 years as of 1 July 2009 (247 840 427) and the CDC estimate for known HIV-positive individuals aged ≥ 13 years (941 950), the annual ED visit rate was estimated as 633 per 1000 persons per year for known HIV-infected persons, while the annual visit rate for HIV-negative persons was 438 per 1000 persons (rate difference 195; 95% CI 194, 197).
Significant differences in age, gender, race, medical bill payment type and location of ED were observed between HIV-positive patient visits and those of the general population (Table 2). Compared with HIV-negative patient visits, a significantly higher proportion of HIV-positive patient visits were by those who were male, aged 45–54 years, Black, publicly insured, and living in metropolitan areas.
ED utilization patterns by HIV status are summarized in Table 3. There were no significant differences between the two groups regarding mode of arrival, whether the visit was the reported initial visit, proportions with abnormal vital signs, acuity level or Glasgow Coma Scale. HIV-positive patients were more likely to have multiple ED visits within the previous 12 months. They were also more likely to have higher pain ratings and more frequent visits for cough or fever. HIV-positive patient visits were associated with significantly longer wait times to be seen by ED providers than visits by HIV-negative patients [mean 76.7 min (95% CI 64.7, 88.7 min) vs. 55.5 min (95% CI 51.8, 59.2 min), respectively]. Significant differences were observed between the two groups regarding the type of provider who evaluated the patient; for example, a much higher proportion of HIV-positive patient visits were seen by any physicians, and residents or interns in addition to attending physicians. HIV-infected patients were also more likely to be seen by consulting physicians [13.7% (95% CI 8.7, 18.6%) vs. 7.6% (95% CI 6.2, 8.9%) for non-HIV-infected patients]. Regarding diagnostic and screening tests and procedures performed in the ED, HIV-positive patient visits were associated with a significantly higher number of any diagnostic test ordered than HIV-negative patient visits [mean 4.5 tests (95% CI 3.9, 5.1) vs. 3.5 tests (95% CI 3.3, 3.6), respectively]. Specifically, a higher proportion of HIV-positive patient visits had a complete blood count, measurements of blood urea nitrogen/creatinine, electrolytes, glucose, cardiac enzymes and liver function, blood cultures and urinalysis than HIV-negative patient visits (Table 3). However, there were no differences between the two groups regarding imaging and procedures performed in the ED. Regarding medications administered in the ED, HIV-positive patients were prescribed and administrated a significantly higher number of medications than HIV-negative patients [prescribed: mean 2.7 medications (95% CI 2.3, 3.0) vs. 2.2 medications (95% CI 2.1, 2.3), respectively; administrated: 1.9 medications (95% CI 1.6, 2.2) vs. 1.4 medications (95% CI 1.4, 1.5), respectively]. Finally, the overall ED duration of stay was significantly higher for HIV-positive patient visits than for HIV-negative patient visits [mean 5.4 h (95% CI 4.6, 6.2 h) vs. 3.6 h (95% CI 3.5, 3.8 h), respectively].
While there were few differences observed in the reason for visits between the two groups, the primary ED diagnosis varied by HIV status. The five leading primary ED diagnoses for HIV-positive patient visits consisted of the following: symptoms, signs and ill-defined conditions (24.4%); injuries and poisonings (12.3%); respiratory system diseases (9.2%); genitourinary diseases (8.3%); and mental disorders (9.3%) (Table 4). The most frequent primary diagnosis received by the HIV-negative population was in the category of injuries and poisonings, the frequency of which was found to be almost twice that observed in the HIV-positive population (21.6% vs. 12.3%, respectively). ED providers diagnosed infectious/parasitic diseases significantly more often in HIV-positive patient visits (6.7% vs. 2.1% in non-HIV-positive patient visits; P < 0.05), and tended towards a higher frequency of diagnosis of mental disorders (7.2% vs. 4.1%, respectively; 0.05 < P < 0.1).
Admission rates were significantly higher for HIV-positive patient visits compared with HIV-negative patient visits [28.3% (95% CI 22.2, 34.4%) vs. 15.0% (95% CI 13.8, 16.1%), respectively]. Multivariate regression analysis showed that, at ED visits, HIV-positive patients were 2.17 times (95% CI 1.28, 2.80) more likely to be admitted than HIV-negative patients, after controlling for other covariates (Table 5). Other significant covariates included age < 55 years, public insurance [odds ratio (OR) 1.43; 95% CI 1.30, 1.57], arrival by ambulance (OR 2.17; 95% CI 1.97, 2.39), visits triaged to an immediate/emergent acuity level (OR 2.01; 95% CI 1.80, 2.24), fever as a reason for visit (OR 1.45; 95% CI 1.17, 1.79), seen by a consulting physician (OR 5.40; 95% CI 4.39, 6.65), seen by a resident/intern (OR 1.41; 95% CI 1.07, 1.86), ordering any blood test (OR 4.80; 95% CI 3.60, 6.41) and ordering of intravenous fluids (OR 2.30; 95% CI 2.03, 2.60).
Results
The 2009 and 2010 NHAMCS surveys together estimated that there were 217 261 467 ED visits (95% CI 195 109 524, 239 413 410) by individuals aged ≥ 13 years in the USA. Of these patient ED visits, 1 192 535 (95% CI 934 410, 1 450 660) were documented to be visits by HIV-infected patients, corresponding to a visit proportion of 5 in 1000 ED visits (95% CI 4, 7 per 1000 visits). Numbers of ED visits and their corresponding visit proportions of all ED visits for known HIV-infected individuals by demographic characteristics are summarized in Table 1. The highest proportions of ED visits for HIV-infected individuals were found for those who were aged 45–54 years, were male, were of Black race, had public medical insurance as payment type, and were living in metropolitan areas and the northeast region. Using the US Census estimate of the civilian noninstitutionalized population aged ≥ 13 years as of 1 July 2009 (247 840 427) and the CDC estimate for known HIV-positive individuals aged ≥ 13 years (941 950), the annual ED visit rate was estimated as 633 per 1000 persons per year for known HIV-infected persons, while the annual visit rate for HIV-negative persons was 438 per 1000 persons (rate difference 195; 95% CI 194, 197).
Significant differences in age, gender, race, medical bill payment type and location of ED were observed between HIV-positive patient visits and those of the general population (Table 2). Compared with HIV-negative patient visits, a significantly higher proportion of HIV-positive patient visits were by those who were male, aged 45–54 years, Black, publicly insured, and living in metropolitan areas.
ED utilization patterns by HIV status are summarized in Table 3. There were no significant differences between the two groups regarding mode of arrival, whether the visit was the reported initial visit, proportions with abnormal vital signs, acuity level or Glasgow Coma Scale. HIV-positive patients were more likely to have multiple ED visits within the previous 12 months. They were also more likely to have higher pain ratings and more frequent visits for cough or fever. HIV-positive patient visits were associated with significantly longer wait times to be seen by ED providers than visits by HIV-negative patients [mean 76.7 min (95% CI 64.7, 88.7 min) vs. 55.5 min (95% CI 51.8, 59.2 min), respectively]. Significant differences were observed between the two groups regarding the type of provider who evaluated the patient; for example, a much higher proportion of HIV-positive patient visits were seen by any physicians, and residents or interns in addition to attending physicians. HIV-infected patients were also more likely to be seen by consulting physicians [13.7% (95% CI 8.7, 18.6%) vs. 7.6% (95% CI 6.2, 8.9%) for non-HIV-infected patients]. Regarding diagnostic and screening tests and procedures performed in the ED, HIV-positive patient visits were associated with a significantly higher number of any diagnostic test ordered than HIV-negative patient visits [mean 4.5 tests (95% CI 3.9, 5.1) vs. 3.5 tests (95% CI 3.3, 3.6), respectively]. Specifically, a higher proportion of HIV-positive patient visits had a complete blood count, measurements of blood urea nitrogen/creatinine, electrolytes, glucose, cardiac enzymes and liver function, blood cultures and urinalysis than HIV-negative patient visits (Table 3). However, there were no differences between the two groups regarding imaging and procedures performed in the ED. Regarding medications administered in the ED, HIV-positive patients were prescribed and administrated a significantly higher number of medications than HIV-negative patients [prescribed: mean 2.7 medications (95% CI 2.3, 3.0) vs. 2.2 medications (95% CI 2.1, 2.3), respectively; administrated: 1.9 medications (95% CI 1.6, 2.2) vs. 1.4 medications (95% CI 1.4, 1.5), respectively]. Finally, the overall ED duration of stay was significantly higher for HIV-positive patient visits than for HIV-negative patient visits [mean 5.4 h (95% CI 4.6, 6.2 h) vs. 3.6 h (95% CI 3.5, 3.8 h), respectively].
While there were few differences observed in the reason for visits between the two groups, the primary ED diagnosis varied by HIV status. The five leading primary ED diagnoses for HIV-positive patient visits consisted of the following: symptoms, signs and ill-defined conditions (24.4%); injuries and poisonings (12.3%); respiratory system diseases (9.2%); genitourinary diseases (8.3%); and mental disorders (9.3%) (Table 4). The most frequent primary diagnosis received by the HIV-negative population was in the category of injuries and poisonings, the frequency of which was found to be almost twice that observed in the HIV-positive population (21.6% vs. 12.3%, respectively). ED providers diagnosed infectious/parasitic diseases significantly more often in HIV-positive patient visits (6.7% vs. 2.1% in non-HIV-positive patient visits; P < 0.05), and tended towards a higher frequency of diagnosis of mental disorders (7.2% vs. 4.1%, respectively; 0.05 < P < 0.1).
Admission rates were significantly higher for HIV-positive patient visits compared with HIV-negative patient visits [28.3% (95% CI 22.2, 34.4%) vs. 15.0% (95% CI 13.8, 16.1%), respectively]. Multivariate regression analysis showed that, at ED visits, HIV-positive patients were 2.17 times (95% CI 1.28, 2.80) more likely to be admitted than HIV-negative patients, after controlling for other covariates (Table 5). Other significant covariates included age < 55 years, public insurance [odds ratio (OR) 1.43; 95% CI 1.30, 1.57], arrival by ambulance (OR 2.17; 95% CI 1.97, 2.39), visits triaged to an immediate/emergent acuity level (OR 2.01; 95% CI 1.80, 2.24), fever as a reason for visit (OR 1.45; 95% CI 1.17, 1.79), seen by a consulting physician (OR 5.40; 95% CI 4.39, 6.65), seen by a resident/intern (OR 1.41; 95% CI 1.07, 1.86), ordering any blood test (OR 4.80; 95% CI 3.60, 6.41) and ordering of intravenous fluids (OR 2.30; 95% CI 2.03, 2.60).
Source...